Hindawi Publishing Corporation
Advances in Mechanical Engineering
Article ID 128432
Research Article
Research on Fault Diagnostic System in CVT Based on UDS
Jiande Wang, Yunshan Zhou, and Quan Li
Engineering Research Centre of Automotive Electronics and Control Technology, Ministry of Education, Hunan University,
Changsha 410082, China
Correspondence should be addressed to Jiande Wang; wjdcvt@163.com
Received 9 July 2014; Revised 6 September 2014; Accepted 15 September 2014
Academic Editor: Hongwei Guo
Copyright © Jiande Wang et al. This is an open access article distributed under the Creative Commons Attribution License, which
permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
A communication model of diagnostic network and implementation of unified diagnostic services (UDS) based on controller area
network (CAN) bus are presented in this paper, and fault diagnostic function of transmission control unit (TCU), USB- (universal
serial bus-) CAN hardware and software modules, and fault diagnostic software based on personal computer (PC) are designed.
Model diagnostic method is applied on ratio control, and fault diagnostic system is tested in vehicle.
1. Introduction
More and more automotive electronic control modules start
to use CAN instead of traditional K (ISO14230) diagnostic
mode due to its higher speed and reliable method. In order
to unify different on-board network diagnostic services, ISO
established a general diagnostic communication protocol,
ISO14229, which is also called UDS. The protocol can
satisfy the demands of Europe on-board diagnostics (EOBD) system and conform to the development trend of
modern network bus system, which will become the future
vehicle industry general diagnostic criteria [1, 2]. Continuously variable transmission (CVT) electronic control unit,
as an important module of automotive electronics, uses a
diagnostic method to monitor the transmission system and
components operation. On the one hand, if a fault happens
in system, the diagnostic system freezes frame data and
provides real-time warning. Also, the diagnostic equipment
can provide real-time, accurate working data, and diagnostic
information for after-sale diagnostic analysis, achieving the
purpose of repairing and monitoring [3].
2. The Communication Mode of
Diagnostic Network
Diagnostic protocol is divided into application layer, network
layer, data link layer, and physical layer. Service
identifier
and the format of the parameters and content of specific
diagnostic service which follow ISO14229 are defined on
application layer. The format of the parameters and content of
specific diagnostic service in service identifier in
application layer conform to ISO14229. Data is transmitted
in the form of a single frame or multi frame, which
follow ISO15765- 2 [4]. On data link layer, it is responsible
for the single frame or multi frame under the network layer
data sent by the physical layer, or which received CAN
frame from the physical layer and submitting it to the
network layer [5]. Diagnostic process is always to be
launched by diagnostic equipment, after electronic control
unit accepting the request of service and resolving step by
step according to diagnostic protocol. Eventually, the
service is passed to the application layer, as shown in
Figure 1.
3. The Implementation of
the UDS on CAN Bus
The unsegmented data flow and segmented data flow are
defined in network layer of ISO15765 protocol. The unsegmented data flow consists of a single frame, where single
frame length is no more than 7 under normal addressing
mode or less than 6 under extended addressing mode. The
segmented data flow consists of a message of multi frame
section. The data length is more than 7 under normal
Downloaded from ade.sagepub.com by guest on September 8, 2015
2
Advances in Mechanical Engineering
Tester
Application layer
Network layer
Data link layer
Physical layer
CAN
node
Diagnostic
service
Diagnostic
service
Message
package/unpackage
Message
package/unpackage
Message
tra nsmit/recei ve
Message
tra nsmit/recei ve
CAN
driver
CAN
driver
Response
CAN_H
CAN_L
Request
Figure 1: The communication model of diagnostic network.
addressing mode or more than 6 under extended addressing
mode. The data stream contains of a first frame and multiple
consecutive frames. Two types of message transmission are
shown in Figure 2.
The single frame, first frame, the consecutive frame, and
flow control all contain data domain of 8 bytes. The data
domain consists of protocol control information and data
transmission. The structure of all kinds of CAN messages is
shown in Table 1 [6].
TCU supports the following addressing modes: physical
addressing mode and functional addressing mode. In the
physical addressing mode, all messages are received by
TCU physical address. TCU can perform services which
are supported by the activation diagnostic session when
the physical addressing mode request is activated. In the
functional addressing mode, message is requested when
request information of diagnostic equipment is not in the case
of a particular module.
TCU supports normal and extended diagnostic modes.
When the vehicle is ignited, TCU is initialized into the
normal mode. Meanwhile, the controller functions and diagnostic services are activated, but the specification execution from the customer is not allowed under the extended
mode.
Table 1: Data domain structure.
Frame type
Byte 1
Byte 2 Byte 3 Byte 4–8
Bit 7-4 Bit 3-0
Single frame
0
SF DL SID
data 1–6
First frame
1
FF DL
SID data 1–5
Consecutive frame 2
SN
data 1–7
Flow control
3
FS
BS
STmin
Remarks: SF DL denotes 4-bit single frame data length; SID denotes
service identifier; FF DL denotes 12-bit first frame data length; SN denotes
sequence number; FS denotes data flow state; BS denotes block size; STmin
denotes the minimum time interval.
4. UDS Diagnostic Services
The application of the diagnostic function of UDS is the
diagnostic services. UDS diagnostic services can be organized
into the six categories, including diagnostic and communication management service, data transmission service, data
storage, input/output control service routines, remote wakeup service, and upload/download service [7]. Diagnostic and
communication management service is to define the main
session control service and the communication parameter
setting, which guarantee the normal work during the process
Downloaded from ade.sagepub.com by guest on September 8, 2015
Advances in Mechanical Engineering
3
Sender
Receive
r
Sender
Receiver
Sender
Receive
r
First frame
Single frame
Flow control frame
Flow control frame
Consecutive frame
Consecutive frame
Consecutive frame
Consecutive frame
Consecutive frame
Consecutive frame
(a) Unsegmented message
(b) Segmented message
Figure 2: Unsegmented and segmented message.
Table 2: UDS diagnostic services.
SID
Service
Service name
Function addressing
Normal mode
Extended mode
Diagnostic and communications management
I
I
Physical addressing
Normal mode
Extended mode
I
I
I
I
I
I
I
I
I
I
e
I
I
I
I
I
I
10
Diagnostic session control
11
27
3E
ECU reset service
Security access
Tester present
I
I
I
22
2E
Read data by identifier
Write data by identifier
I
14
19
Clear diagnostic information
Read DTC information
I
I
Data transmission
Stored data
Remarks: I denotes diagnostic supported under any security condition; e denotes diagnostic only supported under unlock security condition.
of diagnostic communication. Data transmission service
mainly monitors the parameters of vehicle in real-time.
Storing data service reads and clears up the ECU internal
fault code for fault diagnosis. Input/output control function
unit controls the input/output of vehicle ECU. Remote
wake-up service activates ECU internal diagnostic routines.
Upload/download service is primarily used in exchanging a
large number of data between diagnostic equipment and the
internal ECU of vehicles. UDS diagnostic services implementation of TCU is shown in Table 2.
500 kbps. The structure of fault diagnostic system is shown in
Figure 3, including TCU fault diagnostic software, the
USB- CAN hardware and software, and implementation of
TCU diagnostic function.
5. The Design of the Fault Diagnostic System
(1) CAN Communication Module. It receives TCU fault
information and the important parameters in the process of
transmission operation, which is based on the diagnostic
communication specification.
According to ISO15765, CAN information ID from TCU
to diagnostic equipment is set to 0x7E9 (standard frame).
CAN physical addressing information ID from diagnostic
equipment to TCU is set to 0x7E1. CAN functional addressing
information ID from diagnostic equipment to TCU is set to
0x7DF. The CAN communication uses the baud rate with
5.1. TCU Fault Diagnostic Software Design. The software
is developed with the platform C++ Builder, which adopts
modular design. The structure of the software system is
shown in Figure 4.
(2) Fault Information Management Database. It builds up
the fault table and stores the fault information and specific
description of the fault codes.
Downloaded from ade.sagepub.com by guest on September 8, 2015
4
Advances in Mechanical Engineering
TCU fault
diagnostic
software
(PC)
USB
CAN
USB-CAN
TCU
Figure 3: Structure diagram of TCU fault diagnostic system.
Fault diagnostic interface
Software
setting
DTC
operation
Data flow
operation
Fault
information
management
database
Fault diagnostic management module
Monitoring
alarm module
CAN communication module
Figure 4: Structure diagram of TCU fault diagnostic software.
Figure 5: Main interface of TCU fault diagnostic system.
(3) Fault Diagnostic Management Module. It analyzes the fault
diagnostic message for fault diagnostic information. Fault
code, description, and other information can be displayed in
the interface of the fault diagnostic software.
(4) Monitoring Alarm Module. When the failures are detected,
they are divided into general failure and severe failure,
according to the severity of the fault. General failure lights
trouble light, while severe failure flashes trouble light.
(5) Fault Diagnostic Interface. It is used to set the
communi- cation port number, baud rate, and diagnostic
data storage paths and show fault codes. Moreover, the data
flow operation is also in fault diagnostic interface, which
provides real-time data flow monitoring function, and
display the important parameters. The main interface of
TCU fault diagnostic system is shown in Figure 5.
5.2. The Hardware Design of USB-CAN. USB-CAN
include MCU, CAN transceiver and USB serial port
module. Its framework is shown in Figure 6.
The controller uses the MC9S08DZ60 because of its
low cost and low power consumption. CAN communication
module adopts the TJA1040 of NXP. The CAN transceiver
has very low electromagnetic radiation and high ability of
resistance to electromagnetic, which is fully compatible with
ISO11898 protocol and supports high communication rate up
to 1 Mbps. The interface between the USB and serial port
module adopts FTDI’s FT232RL, which is applied for the
USB to serial port interface conversion.
5.3. The Software Design of USB-CAN. The software
function of the USB-CAN is to realize the data interaction
between the message on the CAN bus and the USB of PC
which serves a virtual serial port. This is because many
current
Downloaded from ade.sagepub.com by guest on September 8, 2015
Advances in Mechanical Engineering
5
USB-CAN
OBD
interface
CAN
CAN
CAN
tranreceiver
MCU
SCI
USB
converter
USB
PC
diagnostic
software
Figure 6: Framework of USB-CAN.
TCU
Secondary pressure
Oil temperature
Key lock
Analog input
High side onoff output
Shift lock
Reverse
light
P
R
N
Primary valve control
D
S
Manual plus
Digital input
Manual minus
PWM
output
Snow mode
Secondary valve
control Clutch valve
control Lock valve
Brake
control
Ignition
Calibration
CAN
CANH
CANL
Primary speed
Secondary speed
Turbo speed
Frequency
input
CANH
Communication
CAN
CANL
12 V
GND
Power input
Sensor power
output
5V
Figure 7: Principle block diagram of TCU.
notebook makers canceled the serial port. Since serial port
protocol is simple, the interface is easy to be developed.
The virtual serial port is selected to communicate with PC.
Firstly, slave computer receives some serial communication
data with ID, which was sent by diagnostic software from
PC. Then a message is sent to the TCU by MCU and
MCU receives
a reply message from TCU. After that, MCU restructures the
message and sends it to the host PC.
5.4. The Development of TCU Diagnostic Function. The
TCU diagnostic function diagram is shown in Figure 7.
Downloaded from ade.sagepub.com by guest on September 8, 2015
6
Advances in Mechanical Engineering
Start of service
Request SID
supported?
Negative response
$11 Service not
supported
No
Yes
Session supported?
No
Negative response $7F
Service not supported in
current session
No
Negative response $13
Invalid message length
No
Negative response $12
Sub-function not supported
Yes
Length supported?
Yes
Sub-function
supported?
Ye
s
Positive response
ready?
No
Extended response
time window?
Ye
s
Ye
s
Positive response
Negative response
$XX
No
Negative response $78
Request received
Response pending
No
Other negative
response
End of service
Figure 8: The process of TCU responding diagnostic service.
(1) TCU Self-Diagnosis Function. The input and output
signal have a certain range when CVT is operating. If a
signal value is beyond this range and this phenomenon does
not disappear after a period of time, TCU will make a judge
based on the rule of failure [8]. TCU will store the fault in
the form of a code in EEPROM, lighting trouble light at the
same time.
(2) TCU Failure Protection Function. When TCU detects
some malfunctioning in the sensors or actuators, TCU will
provide alternative signals or setting goals to take the place
of the fault signal, which is to maintain the control system to
ensure the transmission operating.
The flow chart of diagnostic services is shown in Figure
8. The diagnostic services adopt the judgment step by
step. In Table 3, the subfunction, the request parameters,
positive
response identifier, positive response parameters, and the
negative response code in diagnostic services are listed.
Taking diagnostic model control as an example, if diagnostic
equipment request message is 02 10 03 00 00 00 00 00,
TCU
response message will be 02 05 03 00 00 00 00 00. Negative
response message of unmatched session is 03 7F 10 22 00 00
00 00.
5.5. Studying the Method of CVT Fault Diagnosis. The
schematic diagram of CVT hydraulic system is shown in
Figure 9. In CVT, the clamping force of the primary pulley and the secondary pulley is controlled by a hydraulic
valve. The secondary cylinder pressure is controlled by the
proportional relief valve, preventing large metal belt slip
Downloaded from ade.sagepub.com by guest on September 8, 2015
Advances in Mechanical Engineering
7
ẋy
Primary pulley
V-belt
Pump
PP
Ps
Clamping
control
Ratio
control
Secondary pulley
Figure 9: CVT hydraulic principle diagram.
Table 3: UDS diagnostic service definition.
SID
Request subfunction
Request parameter
Positive response identifier
10
01—Normal mode
03—Extended mode
—
50
01—Hardware reset
03—Software reset
01—Request seed
02—Transmit key
—
51
—
xxxxxxxx
67
11
Positive response
parameter
01
03
01
03
xxxxxxxx
—
Negative response code1
12, 13, 22
12, 13, 22, 33
3E
00—Need response
—
7E
0
12, 13, 22, 24, 31,
35,
36, 37
12, 13
22
—
DID2
62
DID + content
12, 22, 31, 33
2E
—
DID + content
6E
DID
12, 22, 31, 33, 72
14
—
01—Read DTC3 number
by identifier
0A—Read all DTC
FF FF 00
54
—
12, 22, 31
59
DTCNo
12, 13, 31
27
19
FF
—
DTC
Remarks: 1negative response code: definition as shown in Table 4; 2DID: data identifier; 3DTC: diagnostic trouble code.
Table 4: Negative response code.
NRC
12 h
13 h
22 h
24 h
31 h
33 h
35 h
36 h
37 h
71 h
72 h
7F h
Meaning
Subfunction not supported
Incorrect message length or invalid format
Conditions not correct
Request sequence error
Request out of range
Security access denied
Invalid key
Exceed number of attempts
Required time delay not expired
Transfer data suspended
General programming failure
Service not supported in active session
between the pulley and the metal belt and also maintaining an effective torque transmission of the engine [9, 10].
In the meanwhile, the primary cylinder pressure is controlled by the proportional reducing valve to achieve ratio
control.
Given the metal belt constraints, there is a nonlinear
relationship between the speed ratio and the displacement
between the primary pulley and secondary pulley. The real
ratio is calculated by measuring the speed of the primary
and the secondary pulley according to the ratio control.
Considering the installation of a shift sensor, cost, and other
factors, the ratio control is achieved by wheel speed
detection of the primary pulley and secondary pulley through
the speed sensor [11].
A ratio control model is designed as shown in Figure 10.
The target ratio is obtained by a look-up table of the throttle
Downloaded from ade.sagepub.com by guest on September 8, 2015
8
Advances in Mechanical Engineering
Fuzzy
control rule
de/dt
Fuzzy
decision
Fuzzy
Ambiguity
Kp Ki Kd
�
d�/dt
Target ratio
+
Ratio
revise
PID
Ratio
control
vavle
PWM
CVT
variator
�P �s
−
V
Real ratio
i = �P/�s
Figure 10: Ratio control model.
2.6
2.4
2.2
Ratio (—)
2
1.8
1.6
1.4
1.2
1
0.8
80
100
120
140
160
Time (s)
180
200
220
Target ratio
Real ratio
Figure 11: Ratio comparing.
position degree � and the vehicle speed �, and then the
target ratio is revised according to the rate of the throttle
changing [12, 13]. Based on the difference between current
ratio and the target ratio, the fuzzy controller calculates PWM
(pulse width modulate) output, controlling the proportional
valve in order to regulate the hydraulic cylinder to achieve
the ratio control [14, 15].
Without the addition of CVT sensors, the online fault
diagnostic ways of model-based ratio control system is
explored [16, 17]. Firstly, the global and local output prediction models are developed based on the collection of
data. Based on the discrepancy between the predicted output
of the model and the actual output, diagnostic services
are performed. Then the failure is isolated and positioned,
according to measuring signal, output results of the global
and local element’s diagnostic model, and the preestablished
diagnostic logic table [18].
6. Diagnostic Function Tests
In the early development stage of UDS function and TCU
fault diagnostic software, the diagnostic function of
Vector’s CAN bus analysis tool CANalyzer is used. The
TCU and laptop are connected by a USB-CAN module for
testing and verifying completeness of whole diagnostic
system. As can be seen in red oval of Figure 11, in the real
vehicle test, the real ratio cannot follow the target ratio
well. So the diagnostic function is activated. Test results of
reading the fault code and data flow in TCU fault diagnostic
software are shown in Figure 12.
7. Conclusion
A TCU fault diagnostic system is developed based on UDS
in this paper. An accurate, fast, stable, and reliable data
Downloaded from ade.sagepub.com by guest on September 8, 2015
Advances in Mechanical Engineering
9
(a) DTC operation
(b) Data flow operation
Figure 12: The test case of TCU fault diagnostic software.
communication between the TCU and diagnostic platform
can be seen in this system. The diagnostic software is
simple and intelligent. The entire system can run smoothly
and diag- nose accurately. Application result shows that the
scheme of the design is reasonable and the system has the
advantages of stable operation, low hardware cost, and a
simple operation. Moreover, it has practical application
promotion.
[9]
[10]
Conflict of Interests
[11]
The authors declare that there is no conflict of interests
regarding the publication of this paper.
[12]
Acknowledgment
The work described in this paper was supported by the
National High Technology Research and Development Program of China (863) (no. 2012AA111710).
References
[13]
[14]
[1] K. H. Johansson, M. To¨rngren, and L. Nielsen, “Vehicle
appli- cations of controller area network,” in Handbook of
Networked and Embedded Control Systems Control
Engineering, pp. 741– 765, 2005.
[2] M. Di Natale, H. Zeng, P. Giusto, and A. Ghosal,
Understanding and Using the Controller Area Network
Communication Protocol, Theory an d Practice, Springer, New
York, NY, USA, 2012.
[3] R. Isermann, “Diagnosis methods for electronic controlled
vehicles,” Vehicle System Dynamics, vol. 36, no. 2-3, pp. 77–
117, 2001.
[4] ISO15765-2:2004, Road Vehicles—Diagnostics on Controller
Area Networks (CAN)—Part 2: Network Layer services,
2004.
[5] M. Rings and P. Phillips, “Adding unified diagnostic services
over CAN to an HIL test system,” SAE Technical Paper
2011- 01-0454, 2011.
[6] ISO15765-3-2004, Road Vehicles—Diagnostics on Controller
Area Networks (CAN)−Part 3: Implementation of Unified
Diag- nostic Services (UDS on CAN), 2004.
[7] ISO, “Road vehicles—unified diagnostic services (UDS)—
part 1: specification and requirements,” ISO 14229-1:2013,
2013.
[8] F. M. Discenzo, P. J. Unsworth, K. A. Loparo, and H. O.
Marcy, “Intelligent motor provides enhanced diagnostics and
control
[15]
[16]
[17]
[18]
for next generation manufacturing systems,” Computing and
Control Engineering Journal, vol. 11, no. 5, pp. 228–233,
2000.
D. Patel, J. Ely, and M. Overson, “CVT drive research study,”
SAE Technical Paper 2005-01-1459, 2005.
M. Pesgens, B. Vroemen, B. Stouten, F. Veldpaus, and M.
Steinbuch, “Control of a hydraulically actuated continuously
variable transmission,” Vehicle System Dynamics, vol. 44, no.
5,
pp. 387–406, 2006.
W. Ryu and H. Kim, “CVT ratio control with consideration of
CVT system loss,” International Journal of Automotive
Technol- ogy, vol. 9, no. 4, pp. 459–465, 2008.
H. Yeo, C. H. Song, C. S. Kim, and H. S. Kim, “Hardware in
the loop simulation of hybrid vehicle for optimal engine
operation by cvt ratio control,” International Journal of
Automotive Tech- nology, vol. 5, no. 3, pp. 201–208, 2004.
Y.-H. Bae, S.-H. Lee, H.-C. Kim, B.-R. Lee, J. Jang, and
J.
Lee, “A real-time intelligent multiple fault diagnostic
system,” International Journal of Advanced Manufacturing
Technology, vol. 29, no. 5, pp. 590–597, 2006.
H. Jin and A. Ge, “On the intelligent slope shift strategy,”
Proceedings of the Institution of Mechanical Engineers Part
D: Journal of Automobile Engineering, vol. 221, no. 8, pp.
991–999, 2007.
L. A. M. Riascos, M. G. Simoes, and P. E. Miyagi, “On-line
fault diagnostic system for proton exchange membrane fuel
cells,” Journal of Power Sources, vol. 175, no. 1, pp. 419–
429, 2008.
B. M. Wilamowski, “Neural networks and fuzzy systems,” in
Mechatronics Handbook, R. R. Bishop, Ed., pp. 1–50, CRC
Press, 2002.
W. Ryu, J. Nam, Y. Lee, and H. Kim, “Model based control
for a pressure control type CVT,” International Journal of
Vehicle Design, vol. 39, no. 3, pp. 175–188, 2005.
A. Scacchioli, G. Rizzoni, M. A. Salman, W. Li, S. Onori, and
X. Zhang, “Model-based diagnosis of an automotive electric
power generation and storage system,” IEEE Transactions on
Systems, Man and Cybernetics, vol. 44, no. 1, pp. 72–85,
2013.
Downloaded from ade.sagepub.com by guest on September 8, 2015